论文部分内容阅读
在机械加工过程,为了提高加工稳定性和精度,在线状态监测具有十分重要的作用。基于经验模态分解与神经网络模型,提出了一个在线状态监测方法。该方法将EMD分解的本征模态函数均方根作为机械加工特征量。为识别实时加工状态,以加工特征为神经网络的目标输入,建立起将IMF作为特征参数及把3种加工状态作为输出的3层后向神经网络模型。识别的结果显示,提出的方法能有效地识别加工状态。
In the machining process, in order to improve the processing stability and accuracy, on-line condition monitoring has a very important role. Based on empirical mode decomposition and neural network model, an online condition monitoring method is proposed. The method uses the root mean square of the intrinsic mode function of EMD decomposition as the machining feature quantity. In order to identify the real-time machining status, a three-layer neural network model with IMF as the characteristic parameter and three kinds of machining states as the output was established. The recognition results show that the proposed method can effectively identify the processing status.